Node Classification 벤치마크
Node Classification on Amazon-Fraud
AUC-ROC
- 2020-08-19 — CARE-GNN: AUC-ROC 89.73
- 2021-04-16 — RioGNN: AUC-ROC 96.19
- 2021-06-18 — RLC-GNN: AUC-ROC 97.48
- 2024-10-21 — LEX-GNN: AUC-ROC 97.91
| Rank | Model | AUC-ROC | Paper | Code | Year |
|---|---|---|---|---|---|
| 1 | LEX-GNN | 97.91 | LEX-GNN: Label-Exploring Graph Neural Network for Accurate Fraud Detection | wdhyun/LEX-GNN | 2024 |
| 2 | GTAN | 97.50 | Semi-supervised Credit Card Fraud Detection via Attribute-Driven Graph Representation | ai4risk/antifraud · finint/antifraud | 2024 |
| 3 | RLC-GNN | 97.48 | RLC-GNN: An Improved Deep Architecture for Spatial-Based Graph Neural Network with Application to Fraud Detection | 2021 | |
| 4 | RioGNN | 96.19 | Reinforced Neighborhood Selection Guided Multi-Relational Graph Neural Networks | safe-graph/RioGNN | 2021 |
| 5 | PC-GNN | 95.86 | Pick and Choose: A GNN-based Imbalanced Learning Approach for Fraud Detection | PonderLY/PC-GNN | 2021 |
| 6 | CARE-GNN | 89.73 | Enhancing Graph Neural Network-based Fraud Detectors against Camouflaged Fraudsters | dmlc/dgl · safe-graph/DGFraud · YingtongDou/CARE-GNN · +3 | 2020 |